Senior AI Platform Architect (Remote)

Insight Softmax Consulting; United States; Remote; USD 180,000 - 240,000; Full time

Job summary

Insight Softmax Consulting is hiring an AI Architect to lead discovery, design and delivery of AI infrastructure for customers building or operating AI platforms. You will prototype, validate assumptions, and hand designs to engineers for implementation.

The role requires hands-on experience with GPU-inference platforms, Kubernetes, and cloud/on-prem deployments. Remote work across multiple time zones is expected, with focus on international collaboration.

Qualifications

• At least 8 years in infrastructure, platform or ML engineering.
• 3+ years designing and operating production AI or ML platforms at scale.
• Hands-on deploying LLM inference in production (e.g., vLLM, TensorRT-LLM, llm-d, Triton).
• Deep knowledge of GPU infrastructure and GPU scheduling on Kubernetes.

Responsibilities

• Lead technical discovery with customers and translate goals into AI platform architectures.
• Design inference serving platforms on Kubernetes and bare metal including autoscaling and multi-tenancy.
• Define reference architectures for multi-tenant GPU clouds and on-prem deployments.
• Advise on AI governance: cost visibility, token accounting, data handling, and security posture for LLM services.
• Review implementations and own technical quality of outcomes.
• Represent the company in customer architecture reviews and pre-sales support.

Skills

Infrastructure engineering

ML platform design

Technical leadership

Tools

Kubernetes

Linux

GPU platforms

TensorRT-LLM

Triton

Envoy/Traefik

Job description

Insight Softmax Consulting is hiring an AI Architect to lead discovery, design and delivery of AI infrastructure for customers building or operating AI platforms. You will prototype, validate assumptions, and hand designs to engineers for implementation.

The role requires hands-on experience with GPU-inference platforms, Kubernetes, and cloud/on-prem deployments. Remote work across multiple time zones is expected, with focus on international collaboration.